10 papers
From Arabic Text to Puzzles: LLM-Driven Development of Arabic Educational Crosswords
Kamyar Zeinalipour, Mohamed Zaky Saad, Marco Maggini +1
We present an Arabic crossword puzzle generator from a given text that utilizes advanced language models such as GPT-4-Turbo, GPT-3.5-Turbo and Llama3-8B-Instruct, specifically dev…
Advancing Student Writing Through Automated Syntax Feedback
Kamyar Zeinalipour, Mehak Mehak, Fatemeh Parsamotamed +2
This study underscores the pivotal role of syntax feedback in augmenting the syntactic proficiency of students. Recognizing the challenges faced by learners in mastering syntactic…
Pirates of the RAG: Adaptively Attacking LLMs to Leak Knowledge Bases
Christian Di Maio, Cristian Cosci, Marco Maggini +2
The growing ubiquity of Retrieval-Augmented Generation (RAG) systems in several real-world services triggers severe concerns about their security. A RAG system improves the generat…
Harnessing LLMs for Educational Content-Driven Italian Crossword Generation
Kamyar Zeinalipour, Achille Fusco, Asya Zanollo +2
In this work, we unveil a novel tool for generating Italian crossword puzzles from text, utilizing advanced language models such as GPT-4o, Mistral-7B-Instruct-v0.3, and Llama3-8b-…
SLIMER-IT: Zero-Shot NER on Italian Language
Andrew Zamai, Leonardo Rigutini, Marco Maggini +1
Traditional approaches to Named Entity Recognition (NER) frame the task into a BIO sequence labeling problem. Although these systems often excel in the downstream task at hand, the…
Show Less, Instruct More: Enriching Prompts with Definitions and Guidelines for Zero-Shot NER
Andrew Zamai, Andrea Zugarini, Leonardo Rigutini +2
Recently, several specialized instruction-tuned Large Language Models (LLMs) for Named Entity Recognition (NER) have emerged. Compared to traditional NER approaches, these models h…